Harbiodun0122/OCR-for-Nigerian-Licence-Plate-

Nigeria license plate reader using OCR technology. This project **"OCR for Nigerian Licence plate"** uses YoloV11 for license plate detection and Fast Plate OCR for reading plate text. ## Dataset used The dataset used is comprised of **4365** images of cars with the Nigerian license plate, divided into 70% train, 20% test and 10% validation. Due to the upload size restriction of github, the link to the dataset is provided here. ## Plate detection model The *yolov11s.pt* model was used for detection of license plate due to its high accuracy and moderate computational cost. Click here to check documentation ## OCR model The model used is **fast-plate-ocr**, a lightweight and fast OCR framework for license plate text recognition. Click here for model documentation. ## Project Workflow Data collection --> Data Preprocessing and Augmentation --> Detection model training --> Cropping out detected plate --> Plate OCR ## Output from video